International Alternative Energy Center
International Alternative Energy Center
Home | Profile | Register | Active Topics | Active Polls | Members | Private Messages | Search | FAQ
Username:
Password:
Save Password
Forgot your Password?




 All Forums
 Technical Help and Questions
 Help each other to help yourself grow
 Joule thief
 New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic  

olddawgsrule
Advanced Member


USA
1434 Posts

Posted - September 23 2013 :  17:37:18  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
This is something I'm only starting into
I realize many of you have already been here and probably moved on
It has reached a point that a few of you have mentioned this
And I need to look at this further, better, deeper.

I'm currently running a Joule Ringer right now and like it.
But have been told the Joule Thief will work better.
I really want to see this/Learn this!

There are so many veriations on the internet...

As a basic circuit I found this:
[URL=http://s988.photobucket.com/user/olddawgsrule/media/basicjoulethief_zps0902c977.jpg.html][/URL]

First question: Is this true circuit true?

Skip wire sizes for now, just basic's.
More questions to come.




Old in age, not in mind, so
'Teach me something new'!

Google AdSense

USA
Mountain View


ron_o
Moderator



United Kingdom
1052 Posts

Posted - September 25 2013 :  18:21:34  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

Yes , thats a true circuit for a joule thief ( it's actually a "blocking oscillator" )

there are a few good circuits at the Talking Electronics website , just look for the free e_book on led circuits ( the 10 led of a 9 volt battery is very good : especially if you use a ferrire core instead of the nut and bolt.

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - September 27 2013 :  02:06:31  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
Thank you Ron!

I saw that diagram on the web and really didn't know where it came from or if it was true.

I will re-visit Talking Electronics for it looks like they have quite a bit of info there to read up on.
I'm home much too late tonight (should say this morning) to research now.
Killer project I'm running for my real job, but almost done...



Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

screaminvern
Junior Member



USA
121 Posts

Posted - October 04 2013 :  18:35:51  Show Profile Send screaminvern a Private Message  Reply with Quote
Hey ODR! You'll like this stuff. I still mess around with mine. If you have any pancake coils, the circuit works very nicely with these for wireless energy transmission.

{History does "not" repeat it's self, idiots repeat history.}
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 08 2013 :  18:07:22  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
Screamin' The statement of "I still mess around with mine", tends to lead me towards you've moved on from it..
Maybe just a "fun" project.

I'm very curious about about this.

I have a Joule ringer running my 120v 6watt bench lights and am rather happy with them.
Then I've been told the "thief" is so much better.
I've yet to find a design to run 120v led's, of same value.

I've ordered some parts.. cheap enough to try.
RonO, I've attempted the math (of which I'll write you if wrong...again..)

I know I'm short on info here.
Only because I don't wish others to follow a blind path.
If it works, full diagram will follow.


Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 08 2013 :  18:38:50  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

I shall look forward to seeing your latest design.

Will you be running at 12 volts again ?

For a "true" joule thief circuit to generate 120 volts will require a drive transistor with an operating voltage ( C - E ) of at least 400 volts ( 800 volts would be ideal )

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 09 2013 :  09:14:58  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
The basic design is Xee2's
It's simple, basic and most of all "I can follow it!"
His design lights a 2w led and if this going to replace my 'Ringer',
then I'll require it to light a 6w led.

I'm going through my stash of transistors to locate a 800-1000v.
Looks like my thought wasn't to far off..

His circuit uses a MJE13007 (400v) and I order a few to replicate his and build from there.

What caught my eye (and attention) is the use of a capacitor in the circuit.

To answer your question: His circuit uses 6v source, so I'll start there as well.

More to follow



Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 10 2013 :  14:34:30  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

Thanks for the update . i shall take a look at Xee2's circuits

ron
Go to Top of Page

screaminvern
Junior Member



USA
121 Posts

Posted - October 16 2013 :  16:51:54  Show Profile Send screaminvern a Private Message  Reply with Quote
Not moved on ODR, I just have things that pop into my head from time to time that cause me to get the circuit out. Actually, your project sounds a lot more interesting than my 1.5V joule thief.

{History does "not" repeat it's self, idiots repeat history.}
Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 16 2013 :  17:05:31  Show Profile Send ron_o a Private Message  Reply with Quote
Hi Vern

I've had a look at Xee2's circuits ( has quite a few , actually ) so now i'm just waiting to see what ODR's building.

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 16 2013 :  17:45:48  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
The MJE13007's are here and away I go.

Here's the diagram I'm following;
[URL=http://s988.photobucket.com/user/olddawgsrule/media/JOULET2_zps1aa89fff.png.html][/URL]

The goal I've set is to out-perform my Joule Ringer.

The joule Ringer will light a 6w (EcoSmart LED, bulb specific) at wattage.
The efficiency comes when lighting multiple lights.
There is reduction in lumen as more are added, but 4 lights are exceptable to me and hit a 180% gain.

Now comes the Joule Thief and my first experiment.

I built the circuit as show and it works.
But I need a 6w light, not the 2w shown for my need.
I used my 6w in the experiment.

Value of the resistor needed to change according to the value of input power.
Thus a change to a potentiometer vs the stated resistor.

One to one, same source of power, I find so far that the thief is 166% more efficient (300mW vs 500mW).

Heat and draw on the 'ringer' was an extreme issue.
The thief seems to have it's own issues of resistance, but not a heat issue.. Well, until you go too far...
Don't go for over-drive...

What intrigued me the most (of this circuit) was the way the resistor and the capacitor come into play.
Yes it is two added components to a simple circuit..
Yet they seem to be necessary.
Reduction in available energy to the LED and a nice balancing of energy.

Before I even attempt a balance of resistance/capacitor adjustment..
The next step is to add additional lights.

What will this do when I add another??

Can this efficiency, one to one, be matched as I go to additional lights?

More to come...







Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 16 2013 :  17:49:57  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
Stole this from Dave's Thread..

Just seemed to fit here..
Something that fits what I'm headed for
Something I ned to look at

here is chapter 5:
You must be logged in to see this link.
scroll down to page 72, its right after the classical fuji circuit, there starts the modified one...

Thanks Luc for posting this!



Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 16 2013 :  18:52:10  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

Thank you for the circuit and your update.

I resently purchased some mains powered led spot lamps ( 9 leds in each ) , after opening one up i discovered a voltage dropping circuit / rectifier etc which when removed and the "guts" of the lamp rewired i now find that it runs on 32 volts. I just wondered exactly whats inside your lamp and are you wasting power ?

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 17 2013 :  16:56:04  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
Funny you should ask this..

I have bought EcoSmart LED 6w light bulbs with an understanding that they seem to run best with Joule Thief/Ringer circuits.
I am seeing this as true so far.

One bulb in particular (out of the 6 I now have) runs so much more efficiently.
I'm talking about a third of the power for same lighting.

I have it put a side for dissection.

On to further testing of thief:
I re-wired to my plug strip so I could add additional lights to the circuit.

First I found that the resistance in the plug strip lowered the light output.
Yet the control of output from the thief is steady vs the Ringer.

With the Ringer, each additional light takes more power and lowers the light value.
Thought the value of light reduction and added power is small, it's still there.

With the Thief the value of power input remains constant as additional lights are added and the lower light value is equal.
At least seems so with my crude way of testing.
One more point for the thief..

The Ringer took much modification to run on different voltages.
The thief, so far, only takes a tuning of the resistor (reason I added a potentiometer).

With 4 lights running on the thief (4x6w) I was running at 6w input(slightly under) and within 20% of the light of the Ringer running at same value (trying to be conservative), using 1 light.

I have tried adding resistors/potentiometer to the ringer and have only burned them.
The cap means a lot here.
This is where my attention is headed.








Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 17 2013 :  19:35:56  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

The capacitor is in effect acting as a frequency dependent resistor ( the oscillator frequency will change with load demand ) automatically adjusting the level of feedback.

BTW it will only work with electrolytic capacitors

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 18 2013 :  15:53:03  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
I've only just started into looking at Trimmer's vs Adjustable plate capacitors and I see you just answered the first question.
The plate capacitor would not work.

Ya know... I kinda like to ask the question before you answer it.. LOL!!

Guess I'm on a quest to see if an electrolytic trimmer exists.

I changed out my pot to a better 10 turn unit, but it's only a 500 Ohm pot (and is within the ranges so far).
I'm stopping by my local electronic surplus store tomorrow to see what's in the 'bin'.
I'm looking for a 1K / 10 turn potentiometer.
I want plenty of room to explore!

More to come!




Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 18 2013 :  19:23:14  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

"Guess I'm on a quest to see if an electrolytic trimmer exists." sorry the answer is no , but you can of course connect other electrolytic capacitors in series ( to reduce final value ) or in parallel ( to increase final value ) A word of warning however when wired in parallel the leakage current may become excessive ( depends upon quality of cap ) so if you do get some "strange" results you will know why.

If you don't have any luck with the 1kohm variable just wire a 470 ohm resistor onto one "track" end connection of the 500 ohm variable resistor , this will at least allow you to increase the range of the resistance.

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 19 2013 :  15:19:37  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
It ends up that increasing the capacitor value only leads to increased draw on start-up. Runs at same value of draw, once running.
No increase in light value.
Not the answer I hoped for..

Now reducing the capacitor value has a different effect.
It allows me to start-up at a lower value of resistance, giving me a greater light value and not costing additional battery draw.
I'm not confirmed yet where I want this to be.
My hopes are to integrate this circuit with my 12v system.
So as you read, understand, I'm pushing this to meet 'my' wants.

As designed, for value of power, for light expected, it seems very valid!
Thank you Xee2 for passing this along.
It's just not the light value I'm looking for, on the lights I'm hoping to run.

Getting closer, and see the value of the capacitor in the circuit.
My apologies Ron,, Had to teach myself what you told me already..

Understanding the capacitors function and having the adjustable resistor, the attention now leads to the transformer.
That was a little too easy to make..LOL..

More to come.



Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 19 2013 :  17:00:42  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

No need for apologies ! experimentation is all part of the learning process and who knows what you will find.

The material composition of the toroid will have a marked effect on the inductors properties due mainly to core saturation and to a lesser extent frequency response .

Since your in "experimenting mode" what is the effect off :-

1) adding a resistor between base and emitter ( start at 10,000 ohms and work down )

2) the inductor stores the energy in a magnetic flux which when released ( transistor switches off ) drives the led so larger inductance ( more turns ) equals more energy.
Problem is , a larger inductance requires longer to build up its flux field . So how can you overcome this problem.

3) primary to secondary ratio , is 20 : 2 ( ie 10 to 1 ) the optimum ( clue:- think about the voltage induced into the secondary )

have fun ( and BTW the answers will yield a 12 volt version )

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 20 2013 :  09:37:01  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
"Since your in "experimenting mode" what is the effect off :-

1) adding a resistor between base and emitter ( start at 10,000 ohms and work down )"

I believe I have that now.. Though the connection travels through the battery.
The higher the ohm, the less draw from the battery, resulting in less light produced.
So far I'm seeing good range in the 70 to 200 ohm range.
Clean start-up at 12v's is around 160 ohms.

"2) the inductor stores the energy in a magnetic flux which when released ( transistor switches off ) drives the led so larger inductance ( more turns ) equals more energy.
Problem is , a larger inductance requires longer to build up its flux field . So how can you overcome this problem."

This one is where I'm at right now. My mind is heading towards wire sizing.
As I look at my ringer, I see 2 different wire sizes between primary and secondary.
An easy test for me is to follow that same principle.
I have some 18g insulated around here and should reduce the resistance by more than a fourth of current resistance.

Think I may just hook the 1K pot back-up and be ready..
Then again, maybe the 5K will be a better choice..

"3) primary to secondary ratio , is 20 : 2 ( ie 10 to 1 ) the optimum ( clue:- think about the voltage induced into the secondary )

have fun ( and BTW the answers will yield a 12 volt version )"

I've changed over to the 10 to 1 ratio and am on my 12v battery.
The 20 to 1 made sense with the 6v source.






Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 20 2013 :  09:40:49  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
Latest version of circuit with changes and notes.

[URL=http://s988.photobucket.com/user/olddawgsrule/media/JOULET2_zps19bda490.png.html][/URL]

More changes to come today regarding the toriod..

Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 20 2013 :  16:40:25  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

Have you thought about what actually switches the transistor off.

The toroid , are you using the full "length" with your windings or are the winding opposite each other.

Reducing the resistance of the inductor will improve it's time constant . Just as a point of intrest ... it is better to use many thin wires ( litz style ) twisted together to form a "bigger" gauge , this reduces "skin effect" in the windings

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 21 2013 :  17:44:31  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
"The toroid , are you using the full "length" with your windings"
The answer is no.
This has bothered me...

The rod used in the 'ringer' is covered.
Rod being a straight line has a definite start and end (a line).
The ring has no start or end, it's infinite.
Not completely covered.

So do I create the start and end with my winding by leaving a space or do I create the infinite cycle by filling..
If I create the cycle, then wrapping over makes sense.
If I do not, then spacing from start/end is where the primary should lay.

Then again, is it created by filling with the combination of both..

This ring has me curious/confused/very interested..


"are the winding opposite each other"
The answer is Yes.

For 'S&G's', I had had to try both directions.
Hey what's a transistor or two at this point.. LOL.
Well.. maybe even a pot..

"it is better to use many thin wires ( litz style ) twisted together to form a "bigger" gauge , this reduces "skin effect" in the windings"
I was headed towards solid till you said this..
Thank you!

For 'S&G's', just had to try a litz primary/secondary..
Missed the ratio by quite a bit.. my bad..
New pot in my future!! LOL.. Pretty colors though.
It was a thought on the infinite side of things.





Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 21 2013 :  19:12:09  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

When winding a toroid it is usual to space the turns evenly around the form. So depending upon the number of turns they could be touching or have gaps between them.

So wind the primary ,turns spaced evenly to "cover" the toroid and then wind the secondary over the primary , again making sure that it is covered.

If you imagine that the toroid could be cut and opened up to form a straight rod then both windings would cover the full length of the rod

The idea is to get maximum coupling between primary and secondary windings.

hope that helps

ron
Go to Top of Page

olddawgsrule
Advanced Member



USA
1434 Posts

Posted - October 22 2013 :  16:12:14  Show Profile  Visit olddawgsrule's Homepage Send olddawgsrule a Private Message  Reply with Quote
That makes sense to me and is how the 'ringer' is wound.
My intention is to follow more of the 'ringer' style of winding and use insulated wire over.

With the ringer, the spacing meant a lot between primary and secondary.

I had to try the designed way of winding and being new to rings..
truly didn't know what to expect.


Old in age, not in mind, so
'Teach me something new'!

Go to Top of Page

ron_o
Moderator



United Kingdom
1052 Posts

Posted - October 22 2013 :  18:21:55  Show Profile Send ron_o a Private Message  Reply with Quote
Hi ODR

Yes i agree , build your replication as per the original design , make sure it works and then see if you can amend/improve the design.

ron
Go to Top of Page
  Previous Topic Topic Next Topic  
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
International Alternative Energy Center © 2000-2009 ForumCo.com Go To Top Of Page
This page was generated in 1.22 seconds. Snitz Forums 2000
RSS Feed 1 RSS Feed 2
Powered by ForumCo 2000-2008
TOS - AUP - URA - Privacy Policy
ForumCo Free Blogs and Galleries
Signup for a free forum or Go Banner Free